Loughborough University
Institutional Repository
The shadow economy
This item was submitted to Loughborough University's Institutional Repository
by the/an author.
Citation: FLEMING, M., ROMAN, J. and FARRELL, G., 2000. The shadow
economy. Journal of International Aairs, Spring, 53, 2, 387-409.
Metadata Record:
Publisher:
https://dspace.lboro.ac.uk/2134/642
c Colombia University
Please cite the published version.
This item was submitted to Loughborough’s Institutional Repository by the
author and is made available under the following Creative Commons Licence
conditions.
For the full text of this licence, please go to:
http://creativecommons.org/licenses/by-nc-nd/2.5/
The Shadow Economy
MATTHEW H. FLEMING, JOHN ROMAN,
AND GRAHAM FARRELL*
“Implicit in much of the literature on the shadow
economy is the view that shadow activity is
undesirable... However, it is not certain that all
shadow economic activity should be discouraged.”
I
n seeking to understand the size of the world economy, we are
usually guided by official statistics on output, trade and
investment. But there is a driving force in today’s world market
that has, hitherto, rarely been recognized for its tremendous
economic impact. It is what we will refer to as the shadow economy.
Its oversight in the calculus of the influences on the international
economy is due to the statistical opaqueness that is part and
parcel of criminal and informal transactions.
Economic activity that falls outside the purview of government
accounting is known by various names: shadow, informal, hidden,
black, underground, gray, clandestine, illegal and parallel.1 Implicit
in each, save for informal, is that these economic activities include
conscious efforts to avoid official detection. Indeed, some
unrecorded activities, such as the drug trade, are illicit. Some
activities, such as day labor, however, are legal but may include
an illicit component, such as unrecorded payments. Still others,
such as household production, are wholly legal. Finally, it may be
that unrecorded activities occur in a sector of the economy that
The views expressed in this paper are those of the authors and do not necessarily
represent those of their affiliated institutions. The authors wish to thank Douglas Keh
and Peter Reuter for their learned advice and input.
1
Jim J. Thomas, Informal Economic Activity (Ann Arbor: University of Michigan Press,
1992) p. X; Edgar L. Feige, “Defining and Estimating Underground and Informal
Economies: The New Institutional Economics Approach,” World Development, 18, no. 7
(1990) pp. 989-1002; Friedrich Schneider and Dominik H. Enste, “Shadow Economies:
Size, Causes, and Consequences,” Journal of Economic Literature (forthcoming 2000).
*
Journal of International Affairs, Spring 2000, 53, no. 2. © The Trustees of Columbia
University in the City of New York.
Journal of International Affairs
is simply unregulated, rather than illicit.
As these nuances indicate, simply defining the shadow economy
is a challenge. Although most authors agree that it is a “fuzzy concept,”2
the notion that activity in the shadow economy has important
implications for the official economy is essentially unchallenged.3 Despite
decades of study on the issue from an empirical and historical point of
view,4 little agreement has been reached on the fundamental constructs
that underlie the shadow economy.
The lack of consensus in formulating a unified theory of the shadow
economy, or even a precise definition of the components that
comprise it, suggests that important questions remain unanswered.
To what extent does the exclusion of shadow economic activity
distort official estimates of macroeconomic variables, including
output, employment and inflation? What are the policy
ramifications of these exclusions? What is the distribution of shadow
economic behavior between unrecorded, but legal, and illicit
activities? Can the overall size of the shadow economy be estimated,
and is it changing over time? Do countries at different stages of
development possess different types of hidden economies? What is
the relationship between regulatory (in)efficiency and the size of
the shadow economy?
This paper gives an overview of issues relating to these questions.
We begin by discussing definitions of shadow economic activity and
why the study of the shadow economy is important. We then review
methodologies employed in the estimation of the size of shadow
economies and present size estimates for several transition,
developing and developed countries. We conclude with thoughts
for future research.
Thomas (1992), p. 333.
Other excellent summaries of the impact of shadow economic activity can be found in:
Schneider and Enste; Hernando de Soto, The Other Path: The Invisible Revolution in the
Third World (New York: Harper and Row, 1989); Douglas Marcoullier and Leslie Young,
“The Black Hole of Graft: The Predatory State and the Informal Economy,” The American
Economic Review 85, no. 3 (1995); Norman Loayza, “The Economics of the Informal
Sector: A Simple Model and Some Empirical Evidence from Latin America,” The World
Bank Policy Research Working Papers Series (February 1997); Carlos Maldanado, “The
Informal Sector: Legalization or Laissez-Faire?” International Labour Review, 134, no. 6
(1995). Also, for a dissent, see Jim J. Thomas who refers to current approaches to
measuring the shadow economy as “measurement without theory”; Jim J. Thomas,
“Quantifying the Black Economy: ‘Measurement Without Theory’ Yet Again?” The
Economic Journal, 109, no. 456 (June 1999), pp. 381-389.
4
Bill Gibson and Bruce Kelley, “A Classical Theory of the Informal Sector,” The Manchester
School, 62 no. 1 (1994) pp. 81.
2
3
388
Matthew H. Fleming, John Roman, and Graham Farrell
DEFINITIONS
Unquestionably, the shadow economy represents some form of
unofficial economy. However, there is extensive disagreement
regarding the definition of the term. To some extent, the shadow
economy is a “(pre)concept in search of a theory,”5 and differences
in the definitions of the shadow economy stem from differing
research objectives, such as estimating the size of the shadow
economy or explaining the motives for shadow economy
participation. In general, there are two approaches to defining
the shadow economy. The first considers shadow economic activity
as simply unrecorded economic activity (the definitional approach),
while the second defines the shadow economy in terms of
behavioral characteristics—its economic activity therein (the
behavioral approach). The former is descriptive while the latter
provides underpinnings of a theoretical explanation for shadow
economic activity. As a result, the components included in these
definitions are necessarily somewhat inconsistent.
A sampling of definitions illustrates these two approaches. Thomas
notes that it is “difficult to provide a formal definition” of the shadow
economy and suggests that it “covers those activities which…are
not recorded in the national income accounts.”6 Schneider and Enste
define the shadow economy in a similar manner as “all economic
activities which contribute to the officially calculated (or observed)
gross national product.”7 Bhattacharyya argues that the hidden
economy is best described as unrecorded national income,
“calculated as the difference between the ‘potential’ national income
for the given currency in circulation and the recorded national
income.”8 Smith defines this sector as “market-based production of
goods and services, whether legal or illegal, which escapes detection
in the official estimates of GDP.”9
The alternative approach finds that shadow economic activities
are best defined with respect to the particular behavioral
characteristics of the activities in question. Feige reports that the
Thomas (1999) p. 387.
ibid., pp. 1- 2.
7
Schneider and Enste, p. 5.
8
Dilip K. Bhattacharyya, “On the Economic Rationale of Estimating the Hidden
Economy,” The Economic Journal, 109, no. 456 (June 1999) p. 348.
9
Philip Smith, “Assessing the Size of the Underground Economy: The Canadian Statistical
Perspectives,” Canadian Economic Observer, Catalogue No.: 11-010, 3.16-33, 18 March
1994, cited in Schneider and Enste, p. 5.
5
6
389
Journal of International Affairs
distinguishing feature between official economic activity and
informal (shadow) activity is:
whether the activity adheres to the established, prevailing
institutional rules of the game…. Adherence to the established
rules constitutes participation in the formal economy…whereas
noncompliance or circumvention of the established rules
constitutes participation in the informal economy.10
Feige adds an important caveat that is generally omitted from
other definitions: “The characteristics of each distinct informal
economy are determined by the particular set of institutional rules
that its members circumvent.” He notes that institutional rules
cause members of the formal sector to “confront a different set
of…transaction costs than those faced by members of the informal
sector.”11 Loayza expands on the argument by drawing on a
definition used by Portes et al and states that the informal
economy is “unregulated by the institutions of society, in a legal
and social environment in which similar activities are regulated.”12
His suggestion is that the shadow economy feeds off “an excessive
regulator y system [that] makes the formal economy
unattractive.”13 Both Loayza and Feige argue that institutional
constraints lead to the formulation of the shadow economy. In
extreme terms, de Soto claims that the shadow economy that
comes about as “[t]he traditional centralism of our society has
proven clearly incapable of satisfying the manifold needs of a
country in transition.”14 Overall, participation in the shadow
economy can be seen as an explicit change in behavior by economic
actors in response to institutional constraints.
Both the definitional and behavioral approaches divide the
shadow economy into four broadly comparable components: the
criminal, irregular, household, and informal sectors. The criminal
sector is defined as illegally produced goods and services, such as
the production and trade of illicit narcotics. The irregular sector is
defined as legally produced goods and services which evade legal
Feige (1990) p. 990.
ibid., p. 990.
12
A. Portes, M. Castells and L.A. Benton, The Informal Economy: Studies in Advanced and Less
Developed Countries (Baltimore: Johns Hopkins University Press, 1989) p. 12; Loayza, p. 1.
13
Loayza, p. 1.
14
de Soto, p. 6.
10
11
390
Matthew H. Fleming, John Roman, and Graham Farrell
reporting requirements, such as tax evasion. The household sector
is defined as household production. And the informal sector is
defined as “economic activities that circumvent the costs and are
excluded from the benefits” 15 of law, such as unregulated
microenterprise.16
It is possible to combine the descriptive and behavioral
approaches to create a definition with shared principles. Definitions
of the shadow economy that include all four components, under
the rubric of exclusion from official estimates of national income or
output, appear to most accurately describe the shadow economy.
Definitions that relate economic activities to regulatory compliance
begin to provide theoretical descriptions of economic behavior. The
two definitions have only subtle differences, related to whether the
definition is used as a basis for formulating an estimate of the size of
the shadow economy (descriptive) or as an explanation for why
shadow activity occurs (behavioral). The underlying theoretical basis
for both approaches appears to be rather weak, so neither definition
yet has substantial predictive powers.
THE IMPORTANCE OF STUDYING THE SHADOW ECONOMY
Why, then, should we be concerned about the shadow economy?
Should we care if transactions occur in the shadow economy rather
than the official economy? Tanzi notes that “there cannot be any
question that the underground economy is a real phenomenon
with important implications that deserve attention and study.”17
Of greatest concern is that this activity is unrecorded, and, as
such, official national income accounts statistics do not accurately
represent the true state of a nation’s economy. Given that these
statistics are employed to generate economic policies, inaccurate
figures may lead to inappropriate policy responses.
At the same time, the emergence and growth of the shadow
economy can also suggest that current policies are misguided.
Indeed, the presence of an active shadow economy often suggests
that existing economic policies, such as tax or regulatory policies,
Feige (1990) p. 992.
These definitions and their attendant labels vary between the definitional and
behavioral approaches, and also with specific regard to the household/unrecorded sector;
see Feige (1990), pp. 990-993 and Thomas (1992) pp. 3-7.
17
Vito Tanzi, “Uses and Abuses of Estimates of the Underground Economy,” The Economic
Journal,109 (June 1999), p. 338.
15
16
391
Journal of International Affairs
are overly burdensome or oppressive. Moreover, the shadow
economy, which is untaxable, reduces potential state revenue. Finally,
while some shadow economic activity is clearly undesirable, such as
illicit trade in narcotics, much of it is generally constructive. It may
provide basic needs, including income, for example, to consumers in
developing countries. Therefore, it is difficult to arrive at an overall
comprehensive assessment of the impact of the shadow economy
on economic growth and development. The following sections discuss
measurement difficulties, policy implications, and the normative
impact of shadow economic activity.
WHY THE SHADOW ECONOMY NEEDS MEASUREMENT
As noted above, the shadow economy is essentially unrecorded,
so official national accounts statistics inaccurately reflect the true
state of the economy. Basic macroeconomic principles dictate that
without accurate accounting, central banks and other monetary
policymakers must establish an effective monetary policy in an
atmosphere of significant uncertainty.18 In other words, effective
monetary and fiscal policy design requires a level of precision in
the estimates of key statistics, such as output and unemployment,
and the presence of non-trivial production in the shadow economy
can distort these measures. Consequently, efforts should be made
to supplement official national accounts statistics with estimates
of shadow economic activity.
The importance of measuring the shadow economy can be
demonstrated by the following example. Several authors maintain
that the intense recession and high unemployment in the late 1970s
and early 1980s in the United States was a mirage,19 and that official
unemployment statistics (which exclude employment in the shadow
economy) missed a substantial shift in employment from the official
to the shadow economy. As a result, monetary and fiscal policy
responses may have been inappropriate. While this is a rather
extreme view of the impact of inaccurate measurement, 20 it
18
J.F. Houston, “The Policy Implications of the Underground Economy,” Journal of
Economics and Business, no. 42 (1990) pp. 27-37.
19
For a description of the processes involved and a review of concurring literature, see
Edgar L. Feige, The Underground Economies: Tax Evasion and Information Distortion
(Cambridge: Cambridge University Press, 1989).
20
Thomas (1999) makes the argument that the proponents of the mirage theory are
mistaken; see pp. 387-388.
392
Matthew H. Fleming, John Roman, and Graham Farrell
highlights the potential importance of measuring the shadow
economy. The impact of these problems is not limited to large
developed nations, such as the United States. Inaccuracies in
national accounts statistics have implications for countries in every
state of development.
Regarding the influence of the criminal sector of the shadow
economy on the official economy, for example, Keh describes the
link between illicitly-derived revenues, spending behavior, and
economic policy development as follows: “spending behavior
becomes influenced not only by the official money supply, but by
the infusion of informal credit as well; in turn, the demand for
money in the official banking system reflects only a part of
domestic economic activity,” and thus, the state has less accurate
information on which to base fiscal and monetary policies.21 While
this statement refers only to the criminal economy, the message is
certainly applicable to non-illicit shadow economic activity.
Inaccurate national accounts statistics may also lead to the
inefficient implementation of social welfare programs. If the
informal sector of the shadow economy is active, for example,
then unemployment may be overestimated and social security
and other welfare benefits may be mistakenly targeted to those
recipients who actually receive some income (albeit from the
informal sector), while those marginally excluded may be in greater
need of assistance. Many observers also argue that there is a direct
relationship between formal and informal markets. Formal
markets may be distorted in the presence of informal economic
activity, with respect to both prices and capital allocation.
Finally, there are political economy implications to inaccurate
national accounts statistics. Official GDP figures are used, among
other things, for several reasons, including: as part of the formula
for allocating quota increases that determine voting rights and
access to credit (and potentially the terms of the credit) in the
IMF, World Bank, and regional development banks; to establish
whether or not countries of the European Union meet various
EU criteria; and, to set the size of members’ contributions to the
EU budget.22 As noted above, the presence of an active shadow
21
Douglas I. Keh, Drug Money in a Changing World: Economic Reform and Criminal Finance,
UNDCP Technical Series no. 4 (October 1996) p. 4.
22
Tanzi (1999) pp. 340-341.
393
Journal of International Affairs
economy renders the official national accounts statistics imprecise,
and incentives may exist for countries to bias estimates of the size
of shadow economies to accord with prevailing political interests.
The importance of accurately representing national accounts
statistics—and the related need for including estimates of shadow
economic activity in such statistics—has not passed unrecognized.
Various international standards for the compilation of national
accounts statistics, including the System of National Accounts
1993, European System of Accounts 1995, General Data
Dissemination System of the IMF, and the Resolution Concerning
Statistics of Employment in the Informal Sector 1993 of the ILO
all call for the inclusion of informal and/or illicit sector estimates,
although they provide little guidance for developing procedures.
Eurostat, in conjunction with Statistics Netherlands, is currently
sponsoring an initiative to examine the feasibility and necessary
methodology of including shadow economy estimates (specifically
illicit activity, such as narcotics) in national accounts statistics.
Yet despite the call for the inclusion of criminal sector estimates,
preliminary study suggests that few EU member states or
candidate countries currently include any shadow economic
activity in their official figures.23 Of course, states must be sure
not to imply that by including estimates of illicit activity in their
national accounts statistics that they are condoning such activity.
IMPLICATIONS
OF
SHADOW ECONOMIC ACTIVITY
There are reasons aside from those related to national accounts
statistics that call for the study of the shadow economy. The
presence of an active shadow economy may reflect the degree to
which various existing economic policies are inappropriate or
inefficient. Oppressive tax and regulatory regimes (or increases in
the tax burden or the degree and/or complexity of regulation over
time) appear to drive economic agents from the official sector
into the unregulated shadow economy.24 As noted by Loayza,
rational actors move to the shadow economy to escape the high
entry costs to legality, such as license fees and registration
Statistics Netherlands, “Minutes, Seminar on Illegal Activities,” (13-14 September
1999) pp. 2-4.
24
Loayza, pp. 3-7; Simon Johnson, Daniel Kaufmannand Pablo Zoido-Lobatón,
“Regulatory Discretion and the Unofficial Economy,” The American Economic Review, no. 88
(May 1998) p. 391; see Schneider and Enste for a survey of the research on this issue.
23
394
Matthew H. Fleming, John Roman, and Graham Farrell
requirements, and the high costs of remaining legal, such as taxes,
red tape, labor and/or environmental regulations.25 The size of
the shadow economy, then, may provide governments with
indicators of potential policy flaws. That said, it is unlikely that
the optimal level of taxation and regulation is that which will
minimize all shadow economic activity.
One role of the state is to provide public goods, such as law and
order—including contract enforceability—funded by tax revenue.
But shadow economic activity escapes taxation, and, as a result,
state tax revenues are lower than they could be. This in turn limits
the ability of the state to provide such public goods and/or the quality
of such goods suffers. There is a recursive component to this dilemma:
as the provision of public goods suffers, fewer agents will have the
incentive to remain in the official sector (not least of all because
those who remain will shoulder an increasing percentage of the
tax burden). As the official sector shrinks, so will government
revenues, driving more agents to the shadow economy. 26
Oppressive tax and/or regulatory regimes which drive agents into
the shadow economy will likely erode the size of tax rolls and thus
hinder the provision of public goods. But even those nonoppressive business or tax environments which have active shadow
economies must face the challenge of providing public goods with
insufficient government revenues. All of this has further
ramifications for the state, because without the rule of law (which
includes the legal enforcement of contract obligations), access to
capital is limited.27
IS SHADOW ECONOMIC ACTIVITY DESIRABLE?
Clearly, then, there is pressing need for study of the size and
nature of the shadow economy. But is all shadow economic activity
undesirable? Should we pursue policies to incorporate all shadow
economic activity into the folds of the official economy?
Implicit in much of the literature on the shadow economy is
the view that shadow activity is undesirable (even the terms
“shadow” or “hidden” carry somewhat negative connotations).
This is most evident with respect to the illicit sector: by making
an activity illegal, a society attaches a stigma to that activity.
25
26
27
Loayza, p. 1.
Johnson, Kaufmann and Zoido-Lobatón, p.387.
Loayza, p. 8.
395
Journal of International Affairs
However, it is not certain that all shadow economic activity should
be discouraged. As discussed earlier, an active shadow economy
might imply the presence of inefficient economic policies. As such,
activity which overcomes inefficiencies in the existing system is
conceivably a good thing: de Soto proposes that non-compliant
activity circumventing onerous regulations may be more
efficient.28 Along these lines, shadow economic activity is thought
to have helped the transition towards market economies and to
have mobilized entrepreneurial endeavors in transition
economies.29 Schneider finds that there is a positive stimulating
effect of shadow economic activity on the official economy, as at
least two-thirds of the income earned in the shadow economy is
immediately spent in the official economy.30 And the informal
sector of the overall shadow economy is recognized in developing
and even transition countries as a provider of fundamental goods
and services.31
From some perspectives it has been argued that even the
criminal sector generates benefits for a nation’s economy. Wealth
generated in the criminal sector may feed back into the formal,
official economy. Maurer, discussing the situation of narcotics in
Colombia, states the following:
...other sectors also profit—indirectly via a multiplier
mechanism—from the employment, income, and demand created
by the drug business. The purchasing power in these sectors has
indirectly created hundreds of other employment opportunities
Much of de Soto’s case is formulated on the concept that Peru is governed by a set of
laws that are relatively less efficient than those that guide the informal sector. For his
description of the consequences of onerous laws, see de Soto pp. 158-172. For a description
of proposed responses to this situation, see pp. 248-259.
29
Aleksandar Štulhofer, “Between Opportunism and Distrust: Socio-Cultural Aspects of
the Underground Economy in Croatia,” in Edgar L. Feige and Katarina Ott, eds.,
Underground Economies in Transition: Unrecorded Activity, Tax Evasion, Corruption and
Organized Crime (Brookfield, VT: Ashgate, 1999) p. 44; Svetlana Glinkina, “Russia’s
Underground Economy During the Transition,” in Edgar L. Feige and Katarina Ott, eds.
Underground Economies in Transition: Unrecorded Activity, Tax Evasion, Corruption and
Organized Crime (Brookfield, VT: Ashgate, 1999) p. 104.
30
Friedrich Schneider, “Stellt das Anwachsen der Schwarzarbeit eine wirtschaftspolitische
Herausforderung dar? Einige Gedanken aus volkswirtschaftlicher Sicht,” Mitteilungen
des Instituts für angewandte Wirtschaftsforschung (IAW), Linz, 98, no. 1, cited in Schneider
and Enste, p. 4.
31
See, for example: International Labor Organization (ILO), Trade Unions and the Informal
Sector: Towards a Comprehensive Strategy, mimeo, 1999, at http://www.ilo.org/public/english/
dialogue/actrav/publ/infsectr.htm.
28
396
Matthew H. Fleming, John Roman, and Graham Farrell
in legal sectors. Capital goods, such as agricultural equipment,
chemicals as well as consumer goods and services (banking,
legal advice) are offered increasingly in the remote coca
cultivation areas. This ‘trickling down’ also represents a
distribution of the drug profits. 32
Such activity comes at a price: the state (or perhaps other states)
faces potentially considerable expenditures in order to prevent
criminal activity, such as police and military counter-drug units
and operations, and society may face higher rates of crime,
particularly violent crime. Also, because criminal activity escapes
taxation, the state must provide the public goods of law and order
with tax revenues that may be insufficient. Benefits, such as those
described by Maurer, need to be weighed in the context of their
political, economic and social costs.
Moreover, we should consider that while shadow economic
activity has in many cases effectively served as a survival
mechanism, or has assisted in the transition to a market economy,
it has not done so without consequences. Glinkina notes that “the
emergence and growth of a large underground economy poses
serious concerns, which on balance dominate” and that:
The effective management of the economy by the state is
undermined. The taxation and foreign exchange base that any
state needs to manage its economy is eroded by the rapid flight
from the official to the underground economy. Macroeconomic
stability is thus harder to attain and sustain. Further, the
legitimacy of the overall legal and regulatory system is
challenged.33
Ott provides an unequivocal assessment, though it is made in
reference to transition economies:
We find that in transition countries, which have a weak
institutional structure for supporting market economic activity,
underground economies pose a grave danger, that of undermining
Eva Maria Maurer, “The Cocaine Business: Impact on Colombia’s Economy and
Possible Solutions,” Eberhard Karls University, Tubingen, Germany (August 1992), cited
in Marion Libreros, “Including Illegal Productive Activities in the National Accounts:
The Case of the Illicit Drug Industry in Some Latin American Countries,” paper presented
to 24th General Conference of the International Association for Research on Income
and Wealth (Lillehammer, Norway, 18-24 August, 1996) p. 14.
33
Glinkina, p. 104.
32
397
Journal of International Affairs
the establishment of the rule of law that is increasingly seen as a
necessary component of successful transition. 34
While informal sector employment may provide employees with
basic subsistence needs, it does so in the face of oppressive and/
or unsafe working conditions; incomes usually at or below the
poverty line; restricted access to state-provided social protection,
training and social services; and frequent exploitation and
infringement of workers’ rights.35
The benefits derived from the shadow economy may or may
not outweigh the costs associated with such activity, and the
influence of shadow economic activity on growth and development
remains unclear.36 This is one area that calls for future research,
particularly with respect to the informal sector, in which the
direction of the impacts are ambiguous.
THE SIZE OF THE SHADOW ECONOMY
Regardless of whether shadow economic activity is viewed as
adverse or benign, there are clearly benefits to understanding the
size and extent of this activity. Considerable work has been devoted
to estimating the size of the shadow economy in countries around
the world, using a variety of methods. Unfortunately, all of these
methods are, to some degree, imperfect, and different methods
appear to generate divergent estimates (for those countries that
are common to multiple studies). The results of such research
have come under intense scrutiny, with some observers suggesting
that without first grounding the concept of the shadow economy
in economic theory, efforts to generate a “magic number” for the
size of the sector are meaningless.37 We briefly discuss the
methodological approaches found in the literature and explore
relevant criticisms. We then present estimates for the size of the
shadow economy in transition, developing and developed
economies.
34
35
36
37
Ott, p. 29.
ILO (1999) p. 1.
Štulhofer, p. 44; Glinkina, p. 104; Thomas (1999) p. 381.
Thomas (1999) pp. 381-389.
398
Matthew H. Fleming, John Roman, and Graham Farrell
Methods of Estimating the Size of Shadow Economic Activity
Various qualitative and quantitative techniques exist to
estimate the size of the shadow economy. In choosing among the
myriad approaches to generating these estimates, we borrow from
Feinstein, who states that “the debate about developing estimates
of illegal or underground economic activity should primarily focus
on practical issues related to how best to go about developing
estimates that are as reliable as possible.”38
Anecdotal and observational approaches to estimating the size
of the shadow economy appear extensively in the literature. These
mainly qualitative approaches draw from information sources
including participant observation, structured and semi-structured
interviews, and generally use single-firm data or surveys to collect
descriptive statistics.39 This qualitative approach is viewed as a
starting point for inquiry and is useful for illustrating relevant
issues and guiding future research.40 In general, the qualitative or
descriptive data used in these analyses are specific to the country
or sector (generally the informal sector) being studied, and can
often provide a wealth of contextual information for developing
broader estimators. However, these studies do not typically
produce externally valid estimates that can be compared to other
studies of the same country, or reliably compared to estimates for
other countries. Therefore, they have limited utility in formulating
a broader understanding of the size of the shadow economy.
The most widely used methods of estimating the size of the
shadow economy take one of two approaches: 1) measurement of
self-reported participation in shadow economic activities; or 2)
measurement of discrepancies between expected levels of demand
or receipts (such as electricity, currency or tax revenue) and actual
use of goods or services.
Census or survey approaches, typically limited to a single
country, have been used to measure the extent to which individuals
self-report (non)compliance with tax laws or unreported work
Jonathan S. Feinstein, “Approaches for Estimating Noncompliance Examples from
Federal Taxation in the United States,” The Economic Journal,109, no. 456 (June 1999)
p. 360.
39
See de Soto for what is generally regarded as the best observational study to date.
40
Feige (1990) p. 993.
38
399
Journal of International Affairs
outside the formal sector.41 While these approaches, in theory,
can capture all components of the shadow economy, their validity
is limited by self-reporting bias and sampling error. These
measurement issues can be overcome, as demonstrated by other
types of self-report studies. In fact, the principal limitation may
be the lack of comparable cross-national data due to the expense
involved in data collection.
Alternatively, the discrepancy approach creates estimates of the
shadow economy by examining discrepancies between reported
activities and unreported activities, or forecasted demand and
actual demand, which are then used to generate estimates of the
overall size of the shadow economy. One discrepancy approach
uses audits of taxpayers to measure individuals’ unreported
taxable income to generate ratios of unreported to reported
income.42 Another similar approach measures differences between
individual expenditures and income or discrepancies between
national expenditures and income.43 These methods have certain
limitations. Estimates may be drawn from a non-representative
sample, such as examining only the behavior of taxpayers, when
the behavior of those who do not pay taxes—those whose principle
incomes are derived in the shadow economy. Or, the models
assume that there is little or no error in the underlying estimates,
which may be an overly-optimistic assumption.
Within the category of discrepancy estimators, two approaches
currently predominate. The “currency demand” method 44
econometrically estimates the expected demand for currency with
no shadow market, and compares this to the actual currency
demand.45 This estimation method is predicated on the notion
See, for example, A. Portes, S. Blitzer, and J. Curtis, “The Urban Informal Sector in
Uruguay: Its Internal Structure, Characteristics and Effects,” World Development, 14
(1986). Also, Schneider and Enste cite a variety of single country surveys.
42
Feige (1990); Edgar L. Feige, “A New Perspective on Macroeconomic Phenomena –
The Theory and Measurement of the Unobserved Sector of the US Economy: Causes,
Consequences and Implications,” presented at the 93rd Annual meeting of the American
Economic Association, 1980; Chris Simon and Ann Witte, Beating the System: The
Underground Economy (Boston: Auburn House Publishing, 1982).
43
Kerrick Macafee, “A Glimpse of the Hidden Economy in the National Accounts of the
United Kingdom in VitoTanzi, ed., ” The Underground Economy in the United States and
Abroad (Lexington, MA: D. C. Heath, 1982) pp. 153-154. For a brief critique of this
approach, see Thomas (1999) p. 382.
44
Tanzi (1983) pp. 77-81.
45
Similar methodology has also been employed in estimating global money laundering
flows.
41
400
Matthew H. Fleming, John Roman, and Graham Farrell
that much of the illicit and informal sector operates in cash to
avoid detection. This method, consequently, excludes non-cash
shadow economic activity. Activities excluded may contain a nontrivial portion of informal or irregular economic activity (such as
bartering). Also, the concept is predicated on the notion that there
is some period during which no shadow economic activity
occurred (this period is used as a baseline for comparison), though
there is little reason to believe that such a shadow economy-free
period exists.46
The other discrepancy approach measures the use of a physical
resource (generally electricity), which is compared to an estimated
expected level of use of that resource at a given GDP level.47 This
approach assumes that the ratio of electricity use and GDP can
be econometrically estimated, and that deviations from expected
levels are attributable to shadow economic activity. However, the
precision of the approach is in question, since it assumes that all
changes in resource consumption are due to shadow economic
activity; the approach ignores potential changes in, for example,
the use of electricity due to technology-driven improved efficiency
in electricity-consuming activities.
Estimates of the size of the shadow economy are imperfect.
This should come as no surprise, due to the difficulty inherent in
estimating activity that is largely carried out by actors who wish
to remain unnoticed and unrecorded. Tanzi notes that “we are
still far from the time when the results of studies of the
underground economy can have immediate consequences for
policy or for the adjustment of various macroeconomic variables”
though he does point out that “the higher the estimates, the
greater should be the incentives on the part of the government to
remove the factors that promote the underground economy.”48
The harshest critic of existing estimation methods to date has
been Thomas, who suggests that they “rely on heroic assumptions
to justify the manipulation of certain numbers” and that the
search for an estimate of the size of the shadow economy as a
percentage of GNP:
Thomas (1999) pp. 382-383.
Alejandro Portes, “The Informal Economy,” in Susan Pozo, ed., Exploring the Underground
Economy: Studies of Illegal and Unreported Activity (Michigan: W.E. Upjohn, Institute for
Employment Research, 1996); Johnson, Kaufmann and Zoido-Lobatón, in Schneider
and Enste.
48
Tanzi (1999) pp. 347, 340.
46
47
401
Journal of International Affairs
...without providing economic theories to explain the
determinants and structure of the black economy has led
economists into a back alley in which the question of size has
become an end in itself and more important issues have not
been addressed.49
Estimates of the Size of Shadow Economic Activity in Transition,
Developing, and Developed Economies
To provide a basis for discussion, we present estimates of shadow
economic activity (grouped by transition, developing, and
developed economies) drawn from Schneider and Enste’s timely
summary of the measures of shadow economic activity around
the world. Acknowledging that estimates of the shadow economy
are imprecise, the level and composition of shadow economic activity
differs—not surprisingly—between and among transition, developing,
and developed countries. Undoubtedly, differences in social,
political and legal characteristics of countries explain to a great
extent the size and composition of shadow economies, and why
shadow economies differ by country type. Although much work
has examined shadow economies—or specific components therein
—in countries at different stages of development, there is a dearth
of comparative work both: 1) between such country groupings; and
2) regarding the distribution of component activities (i.e., criminal,
informal, irregular, and household) between and within overall
shadow economies. Explanations for trends are anecdotal and/or
speculative at this stage. As a result, in the discussion that follows,
we offer very general insights from the literature. Further research
should examine in much greater detail the similarities and
differences of shadow economies between country types and also
between, inter alia, countries with similar legal systems, similar tax
regimes, and/or similar fiscal burdens. Perhaps more importantly,
research should also explore the country-specific distribution of
shadow economic activity between the component sectors.
The precision of the estimates presented below is still under
discussion, and consequently, it may be useful to think not only
about absolute measurements, but about the rank order of the
sizes of shadow economies between countries. Also, as pointestimates, these figures do not capture the dynamic nature of
changes to the size of shadow economies over time, which may be
49
Thomas (1999) pp. 382, 387.
402
Matthew H. Fleming, John Roman, and Graham Farrell
of greater relevance. However, the estimates do give us a reference
point for discussion.
Transition Economies
Estimates of the size of shadow economies in transition
economies vary between seven and forty-three percent of GDP
(see Table 1 below). For transition economies (and perhaps
others), the size of the shadow economy may be an indicator of
the level of economic and political liberalization and the success
of macroeconomic stabilization,50 which suggest that, based on
these figures, the Czech Republic, Romania, and Slovakia are in
the best shape.
Table 1
Estimates of the size of the shadow economy in various
transition economies
Country
Size of shadow economy in percent of
GDP, average over 1990-93
Central Europe
Hungary, Bulgaria and Poland
20-28%
Romania, Slovakia and Czech Republic
7-16%
Former Soviet Union
Georgia, Azerbaijan, Ukraine and Belarus
28-43%
Russia, Lithuania, Latvia and Estonia
20-27%
Source: Schneider and Enste (forthcoming).
In general, shadow economic activity in pre-transition
economies appears to have been driven in part by the need to
circumvent inefficient policies often found in command
economies.51 In the transition to a market economy, a tradition of
shadow economic activity—what Feige refers to as the “legacy of
50
51
Glinkina, p. 103.
Feige and Ott, eds. (1999) p. 4.
403
Journal of International Affairs
noncompliance”—and deficient state mechanisms for establishing
and enforcing the rule of law seem to have reinforced the role of
shadow economies.52 Johnson, Kaufmann, and Zoido-Lobatón
note that:
…in Eastern Europe and the Former Soviet Union since 1989,
businesses have responded to politicization by going
underground. Instead of registering their activities, managers
prefer not to pay taxes and not to benefit from key publicly
provided services, such as legal enforcement of contracts…most
of the former Soviet Union has thus ended up in a ‘bad’
equilibrium with low tax revenue, high unofficial economy as
a percentage of GDP, and low quality of publicly provided
services. 53
Kaufmann and Kaliberda report that unofficial (shadow)
activity in the economies of the former Soviet Union and
Central and Eastern Europe displays the following
characteristics: a coexistence of state and non-state activities
and enterprises in the unofficial economy; considerable visibility
and size of unofficial activities; unofficial and unregulated
activities, which are mainly nonviolent and non-criminal;
activities that exist on a continuum in the official-unofficial
activity spectrum (i.e., there is no sharp dichotomy between
official and unofficial activities); social services and state
subsidies are accessible to unofficial activities (i.e., virtually
every activity and individual is within the social security net);
and the unofficial economy is shallow (i.e., actors move between
the two markets relatively freely, motivated by governmentinduced incentives).54
Not all observers agree, however. Glinkina suggests that at
least in Russia, the composition of the shadow economy is very
much dominated by criminal activity. 55 Taken together, it
appears that the two components of shadow economies in
transitioning economies tend to account for most of the shadow
market: irregular and criminal activities. However, little
52
53
54
55
Feige (1999) pp. 17-18.
Johnson, Kaufmann and Zoido-Lobatón, p. 387.
Kaufmann and Kaliberda, pp. 2-5.
Glinkina, pp. 105-115.
404
Matthew H. Fleming, John Roman, and Graham Farrell
evidence currently suggests that one component is more
responsible for shadow market activities than others.
Developing Economies
The shadow economy in developing countries appears in many
cases to be quite large—much larger than in transition or
developed economies—though the range between countries,
thirteen to seventy-six percent of GDP, is also considerable (see
Table 2 below).
Table 2
Estimates of the size of the shadow economy in various
developing economies
Country
Size of shadow economy in percent of
GDP, average over 1990-93
Africa
Nigeria and Egypt
68-76%
Tunisia and Morocco
39-45%
Central and South America
Guatemala, Mexico, Peru and Panama
40-60%
Chile, Costa Rica, Venezuela, Brazil, Paraguay and Colombia
25-35%
Asia
Thailand
70%
Philippines, Sri Lanka, Malaysia and South Korea
Hong Kong and Singapore
38-50%
13%
Source: Schneider and Enste (forthcoming).
Of course, countries of the developing world comprise a large
and varied group, and there is little reason to believe that their
shadow economies share anything but the most general of
attributes. Some of the most promising work relating to developing
countries has focused primarily on Latin America; moreover, the
literature on shadow economic activity in developing countries
focuses almost exclusively on the informal component of the shadow
economy, particularly the role of the informal sector in
405
Journal of International Affairs
employment. The ability to draw broad conclusions about all
developing countries (and, as noted, about all countries) and the
overall shadow economy is limited.
That said, the informal sector component of the shadow
economy appears to be of considerable importance to the
economic activity of most, if not all, developing countries. The
ILO reports that 80 percent of new jobs created between 199094 in Latin America were in the informal sector. Urban informal
employment absorbs 61 percent of the urban labor force in
Africa, and is expected to generate more than 93 percent of all
additional jobs in the region in the 1990s. The informal sector
also absorbed between 40 and 50 percent of the urban labor
force in pre-crisis Asia. 56 Other observers tend to agree, as
Jagannathan notes:
Markets in [developing countries] do not fit the smoothly
operating institutions assumed by standard neoclassical
economics…Even in the poorest of developing countries, there
are some markets, in what can be described as the organized
sector, which function within the framework of civil laws and
administrative regulations…Outside the pale of the organized
sector is the unorganized sector, or informal sector, covering
economic transactions of anywhere between 30 percent and 70
percent of the country’s GNP…It covers most rural markets and
the peripheral but growing urban informal sector.57
We cannot conclude, however, that the informal sector alone
accounts for the lion’s share of the overall shadow economy in all
developing countries. There is every reason to believe that the
criminal sector, such as cultivation, processing, and distribution
of illicit narcotics; trafficking in goods/people; or, prostitution, is
nontrivial in size in many developing countries. Colombia, for
example, recently announced that it would include income from
growing illegal drug crops in calculations of the nation’s economy,
given the size of such activity.58 Bolivia, Peru and Colombia are
the world’s major producers of cocaine, and Afghanistan, Burma,
Laos, Mexico, Pakistan, Thailand and Colombia are the world’s
ILO (1999).
N. Vijay Jagannathan, Informal Markets in Developing Countries (New York: Oxford
University Press, 1987), p.4.
58
“Colombia Adjusts Economic Figures to Include Its Drug Crops,” New York Times, 27
June 1999.
56
57
406
Matthew H. Fleming, John Roman, and Graham Farrell
largest producers of heroin.59 These countries are most likely to
have active criminal sectors.
The relationship between the shadow economy and the illicit
drug trade is a complex one. It is typically believed that illicit
drug production primarily stems from economic causes. Yet a
preliminary study comparing the existence and extent of illicit
drug cropping to both per capita GDP and the Human
Development Index did not find any significant relationship.60
To conclude that economic development is uniformly the
appropriate response to reduce this area of the shadow economy
may be overly simplistic.
Developed Economies
While the evidence suggests that there is a significant shadow
economy even in developed countries—estimates range from eight
to thirty percent of GDP (see Table 3 below)—the magnitude of
the shadow economy is generally less than in transitioning or
developing countries. Perhaps most surprising, however, is the fact
that these estimates suggest that even in developed economies,
the size of the shadow economy may be noteworthy.
Table 3
Estimates of the size of the shadow economy in various
developed economies
Country
Size of shadow economy in percent of
GDP, average over 1990-93
Greece, Italy, Spain, Portugal and Belgium
24-30%
Sweden, Norway, Denmark, Ireland, France,
The Netherlands, Germany and Great Britain
13-23%
Japan, United States, Austria and Switzerland
8-10%
Source: Schneider and Enste (forthcoming).
Central Intelligence Agency, “Major Coca and Opium Producing Nations: Cultivation
and Production Estimates, 1994-98,” mimeo, 1999, at http://www.odci.gov/cia/di/
products/coca_opium/index.html.
60
Graham Farrell, “A Global Empirical Review of Drug Crop Eradication and United
Nations Crop Substitution and Alternative Development Programmes,” Journal of Drug
Issues, p. 28.
59
407
Journal of International Affairs
The countries of southern Europe lead the pack, most likely
due to what at least anecdotally appears to be the presence of
large irregular sectors.61 Scandinavian countries tend to have high
tax rates, so it may be the intense regulatory and fiscal burden
that drives individuals to the shadow economy.62 Generally, it
seems plausible that informal sector activity is of lesser importance
in developed economies. This perhaps explains, in part, smaller
shadow economy estimates than those for transition/developing
countries.
CONCLUSIONS AND THOUGHTS FOR FUTURE RESEARCH
The shadow economy is an active component of the majority, if
not all, of the world’s economies, though its size appears to vary
considerably between countries. Shadow economic activity should
be included in national accounts statistics, to the extent possible,
given the obviously incomplete picture generated by ignoring such
activity. But it seems unlikely that robust estimators can be
developed without a clear understanding of both: 1) the
underlying economic theory describing how shadow economies
come about; and 2) what drives the level of activity in each
component sector.
It may be time to disaggregate the shadow economy into its
component parts, and to focus research efforts on these sectors
independently. For instance, household production is clearly a
valid component of the shadow economy, but the underlying
economic behavior is radically different than that associated with
the criminal sector behavior of the illicit drug trade. Further, it
appears that there is a natural split in the literature between the
informal sector (specifically unregulated production) and the
criminal sector (including drug trade, money laundering and tax
evasion). Research describing the economic behaviors that drive
these activities would provide an important framework for the
future investigation of the magnitude and consequences of their
effects. Only by studying individual components will we gain a
61
See, for example, Bruno Contini, “The Second Economy of Italy,” in Vito Tanzi, ed.,
The Underground Economy in the United States and Abroad (Lexington, MA: D.C. Heath
and Company, 1982) pp.199-208.
62
Schneider and Enste, p. 8; Ingemar Hansson, “The Underground Economy in a High
Tax Country: The Case of Sweden,” in Vito Tanzi, ed., The Underground Economy in the
United States and Abroad (Lexington, MA: D.C. Heath and Company, 1982) p. 241.
408
Matthew H. Fleming, John Roman, and Graham Farrell
better understanding of them. Among the questions to be
addressed are the following: How might we accurately measure
the size of each of the sectors in specific countries? Can we find
methods which employ accurate, obtainable data? Can we find
methods which allow for the comparison of findings between
countries? What are the impacts of such sector activity on
development and real GDP growth? How might we reduce the
level of activity in those sectors which are undesirable? And what
is the country-specific distribution of shadow economic activity
between the component sectors, and do the sectors have an
impact on each other? In short, while the extent and the
significance of the shadow economy are being increasingly
recognized, knowledge is limited, and our portfolio of appropriate
responses is in its infancy.
409